What is WAFL part V: Confusion and Conclusion

So why the confusion?

The optimist in me believes that the reason we have so much confusion about WAFL and SAN is that NetApp has not done a good job explaining the layering. And that without an explanation, it was reasonable for folks to assuming the following, very incorrect layering:

Or using my object diagram

 

that a File IS-A inode and a LUN IS-A File.

And some concluding thoughts and an apology.

This all began because of a comment I made in a response to a comment by Steven Schwartz. And I said later on, in a rather harsh tone of voice:

Comments, such as yours, that WAFL == FS, remains one of the more entertaining, misleading and ill-informed comments that our competitors continue to make on a regular basis.

In retrospect, if Steven was ill-informed that was my fault, and if he was ill-informed then he could not be misleading, and shame on me for being entertained by his ignorance.

So I apologize for my response, because it was rude.

I hope, in future, this discussion about whether WAFL is or is not a file will not be necessary.

WAFL is not a file system. WAFL provides mechanisms that enable a variety of file systems and technologies that want to access disk blocks. WAFL mechanisms deliver industry leading performance, while offering customers the flexibility to use any and all of our data management primitives with no compromise to that performance.

Comments

[karld]

Hi I've posted a response blog talking about WAFL at http://www.communities.hp.com/online/blogs/datastorage/archive/2008/11/03/making-sense-of-wafl.aspx. Feel free to address anything I've said. Karl

[kostadis roussos]

I gave my response here.

http://blogs.netapp.com/extensible_netapp/2008/11/progress-of-a-k.html

[TimC]

I guess I still don't follow your reasoning. There's a layer internal to the filer that interprets the calls and points them from whatever protocol to an inode. A cifs request isn't coming in asking for a specific block, and an FC request isn't coming in asking to modify a file. The filer has to turn "open file tim.txt" into "go grab this inode".

Furthermore, if a LUN isn't being treated as a file, why is it that the filer has no idea what goes on inside of a LUN? If I write out 20GB of data, then delete 10GB of data, the filer still treats that as 20GB used. If it's truly not a "file" in the sense we traditionally think of it, the filer should be fully aware of just how many of those blocks are currently in use.

For all intents and purposes, the filer is treating a LUN out there like one big "blob" or "file". Heck, if I mount up a cifs or nfs share to that volume I can SEE the "blob", which further strengthens the argument for a LUN being a file.

Per one of the other comments in another post of yours: I don't really follow why any of that matters though. Outside of the fact that people like facebook can do some cool, completely undocumented/unsupported calls via nfs... who cares? It either works, and is fast, or it doesn't.

I think in any case you're going to have a real tough sell with this one, and I'm in your corner.

[kostadis roussos]

TimC,

I think the question about whether a LUN is a file or not is really an interesting philosophical discussion of limited customer value. In my mind, the problem is that the definition of file is too imprecise. I have made my case.

Amazingly, a significant amount of EMC FUD,

http://oraclestorageguy.typepad.com/oraclestorageguy/2008/02/blended-fcp-n-1.html

about our system is devoted to making people think that the layering of SAN on WAFL looks like the pictures I drew in this post.

Just recently HP has taken to pursuing the same line of reasoning.

I have my opinions. Dave has expressed his here:

http://blogs.netapp.com/dave/2008/12/is-wafl-a-files.html

At the end of the day, I believe that what's important is whether we add value to your environment.

As I've said before,

http://blogs.netapp.com/extensible_netapp/group_theory/index.html

architectural orthodoxy is the last defense of the incumbent just before they are walked out to pasture.

cheers,
kostadis